I see there have been many threads over the years about the best way to deal with very large WAV files, such as can come from recording long concerts with a continuous WAV file.  In the past, I have recorded mainly in stereo, and it was pretty easy to manage that in Audacity.  Audacity has a nice feature where you can label the individual songs (time ranges), and then do a single "Export Multiple" command to put each of the songs out as a separate WAV or MP3.  That was a very fast workflow.  But I am moving beyond Audacity to get higher quality effects, the benefits of mixing automation, and the ability to handle more than 2-track recording efficiently.  I expect to do a lot of 4-channel live recordings and some 8-channel live recordings in the next year, so I am looking for the best workflow in Sonar.
 
Many people have been concerned about the speed and hassle of those gigabyte-sized WAVs.  I did a 4-track recording last week, so my first shot at this process was to break the songs into separate WAVs using Audacity, then create a separate Sonar project for each song.  This isn't as bad as it sounds because I ended up with two templates (one for instrumentals and one for songs that had a singer.)  And I could just clone each project file, keep the tracks intact, but delete the old clips and drop in new clips into each track.  If there wasn't some special processing for a song, it took about 5 minutes to edit each song after I got the templates set up.  But still this is a lot of work, and will get much worse with 8 tracks.
 
So now I am headed a different direction.  I should mention that I have some large network-attached storage (NAS) boxes, and that's where I want all the finished projects to reside.  Ideally, I'd like to do the editing from start to finish on the NAS.  That sort-of worked out on the above project because each file wasn't too large.  But that really isn't the right answer if I'm going to process the full program at once in SONAR.  It turns out I had an Intel SSD drive on another machine, where it wasn't really needed, so my first step was to recommission that SSD drive to my Sonar machine.  While the NAS will still be the permanent home, I will do all the editing from the SSD.  That is pretty easy to accomplish because I can do a batch upload to the SSD before I start the project, and I have a replication program that will automatically copy any projects on the SSD to their permanent home on the NAS each night.
 
This seems to make it very practical to handle the big WAVs in Sonar.  Setting up the project takes a little time to read the WAVs into Sonar, but after that, everything seems just as fast as with smaller files.  One quirk is that if you drag and drop a big WAV to (what you think is) measure 1, it actually ends up way off to the right in measure 5000 or something, but it isn't hard to get the WAVs all moved over to the left.  From there, it is the normal process of selecting compression, EQ, reverb, or whatever.  There is a feature in Sonar to link multiple clips together for selection purposes.  This is a must-do.  And then when you use the split tool, each of the split sections retains that linked relationship, which is nice.
 
So I envision setting up the effects for the entire concert, and then doing any tweaking required for each song.  When a song is sounding good, just select that clip and File-Export-Audio.  With the SSD, a 4 minute song exports in under 10 seconds on my system.
 
I haven't checked, but I expect this will not be bad for storage space.  Doing it this way, I think the AUDIO folder should just have one copy of my WAVs. And as a practical matter, I have mostly given up on the idea of using 96 KHz.  I will do 44.1 at 24 bits I think, or maybe just stick with 16 bits.  That will keep the file sizes down.
 
I'm thinking this is the fastest workflow for this kind of project.  Any other ideas out there?

Some great ideas and reflections, thank you.

From my experience with recording live shows in multi-track:

• ALWAYS record in 24-bit. This is important for headroom, sound quality, and avoiding overs. When recording in 16-bit you are losing resolution every moment you don't record at full volume, and risking overs every time you come close to full volume. Especially with live sound, 24-bit is a must.
• Whether you do one project or multiples depends on the kind of show and the output medium. If you're looking to have the output be a complete A to Z recording of the live show (as an archive, perhaps), then one project is the way to go. If you're going to be picking and choosing songs for a podcast or video, then separate ones might work. If there are multiple groups or bands performing, one project per band can work well, too. I usually create a project with the whole waves imported, then do a Save As for each band and cut down the waves so I can focus just on that band
• Choose your plugins carefully. I have a number of UAD plugins but don't have a lot of UAD dsp power, and made the mistake of buying the Studer A800 Tape plugin when it came out. It's an excellent plugin, but with the small amount of UAD dsp I have, it meant I would have to apply or freeze the tape plugin on tracks to clear up DSP so I could use more UAD plugins
• Pick a few reverbs and put them on busses, then send desired tracks to them. This uses less overhead and makes for a more cohesive sound
• There are two relevant options in the Preferences, under File, then Audio Data: "Always Import Broadcast Waves At Their Timestamp" and "Apply SMPTE Offset to Broadcast Wave Timestamps." You may find a combination of these that imports all WAV files to the beginning of the timeline
• As for splitting files, the fast way I've found to do this is to run through the recording, looking for places where songs end, banter happens, songs begin, etc, and I place markers everywhere I want a split. Then you can select all, right click on a clip and hit Split, and use the Split At Each Marker option

I don't know your system specs, but I've handled 30+ tracks at 24/44.1 that were 2-3 hours long in Sonar. Space can be an issue for sure, but with SATA internal drives or modern external options (like FireWire, eSata, USB3) you shouldn't have major issues. As you mentioned, the waveforms take a while to draw when you first import files, and applying FX can take a while when working with long files, but that's about it. (That, and Sonar doesn't allow you to zoom out very far, so it's a bit of a hassle if you try to keep everything in one project and want to see things in context.)

Peace,

Tunes

Thanks so much for those tips.  It is really helpful.  I was thinking the same thing about the importance of 24 bits, but you solidified that for me.  Often I am playing in these concerts, so I have to go with mic placement and recording levels, which are really just guesses based on the sound pressure as the group warms up.  I have to set it conservatively, of course, because if I clip in the recording, all is lost.  24 bits should give me a little more safety factor there.
 
I can't envision doing 30+ tracks in the next few years, so I am comfortable that performance will work out OK with 8 or 12 tracks.
 
This means that I really don't get close-in mics on anything except vocals or instrumental solos (with a clip on wireless mic).  I know that isn't how a top notch live recording is produced, but I actually get pretty good results with area coverage, sometimes using EQ rather aggressively to make up for the smaller number of mics.
 
I should point out these are mostly instrumental groups (jazz bands, concert bands, etc.) so the stage is relatively quiet.

Actually, for what you're doing, close mics would be weird for things other than solo instruments or vocals. It sounds like you have the right approach there.

With that small a number of tracks I probably wouldn't bother splitting into separate projects. I would expect to have two classes of track- for the distance mics I would expect I would do processing that is pretty consistent across the board, and probably wouldn't tweak much. For the solo instruments and vocals I would expect to do a lot more detail work.

Out of curiosity, what kind of mic setup do you use for distance mics? Spaced Omnis? Spaced Cardiods? Coincident pair of cardioids? ORTF Cardoids? Mid-Side? Blumlein? Decca Tree?

Peace,

Tunes

Do it a few times and see what you learn. Using a separate project for tunes or groups of tunes makes it easier to go back and make changes. Bouncing everything out of one project is more efficient.

tunekicker
Out of curiosity, what kind of mic setup do you use for distance mics? Spaced Omnis? Spaced Cardiods? Coincident pair of cardioids? ORTF Cardoids? Mid-Side? Blumlein? Decca Tree?

It is all cardioids, because that's what I have.  Normally, considering these are acoustical groups that are supposed to know how to play in balance, I concentrate on one stereo pair at the front of the ensemble.  This will either be the inboard mics on a Zoom H4n or else a Rode NT4, which sounds just a little better, I think.  Then I look for all the places where I think I might later find myself wishing I could boost a little.
 
For example, one concert band plays some concerts in an open field with no shell.  You lose all the lows, so I try to get a mic in the tubas, and pick up some euphonium that I can cut out with EQ if I don't need it.  The percussion are at the back and generally get lost, so I try to at least capture the drum set.  A single MXL 603S has worked very well placed about 4 feet in front of the drum set.  I know this sounds like sacrilege to anybody doing rock setups, but in a concert band, that gets the essence of the drum sound into the mix.  Saxophones tend to get lost in this particular ensemble as well. 
 
For a jazz big band, I again start with a stereo pair as the main source, but will have extra mics to isolate piano, bass and drums.  Trombones tend not to carry, so I might try to get a couple of mics at about the sax player's "elbows".  That will get trombone and trumpet, but not so much sax, and I can scale back the trumpet with EQ.  Lately I have used a Samson Airline 77 for trombone solos.  This is a handy little wireless mic with no separate battery pack.  It can easily be passed up and down the section to different soloists.
 
I am looking at an Allen Heath ICE16 as a simple way to capture these live channels without the complications/risks of a DAW in the field.
 
 

I've used recording units like the ice with great success. I've loved the akai dps and Tascam 2488. Handy tools!

forum.cakewalk.com/m/tm.aspx?m=2852986

In this thread I detail how I recently made such a recording.

gswitz
I've used recording units like the ice with great success. I've loved the akai dps and Tascam 2488. Handy tools!

The Allen & Heath looks attractive to me because it is line level only on input, so you aren't paying for so-so preamps.  It is about $1000 for 16 channels of recording.  I have a Presonus StudioLive which has DB25 connectors that can fan out to the inputs of an ICE.  I can capture all my live PA mics and then worry about trying to do something with that material later in the studio.

When I'm doing live recording, I rarely launch Sonar unless I need a click-track for the band. Even then, the Tascam can provide a click or even drums and it is much quieter than my quietest laptop.
 
Personally, I think you've got a good plan here.
 
For me, I have really grown over time. I have been able to use the Tascam to automate faders in Sonar (and Sonar to automate faders in the Tascam). You can do all kinds of things, so let your imagination flow as you consider how you might use whatever you buy. The Tascam and the Akai I own both had faders so I can use them for the live mixing console. FX like compression and EQ too. Without this feature, there are lots of cases where I could not have put my gear between the mics and the mains. :-)